The use of contextualised standardised client simulation to develop clinical reasoning in 1
final year veterinary students 2
3 4
Claire EK Vinten; BVMedSci BVM BVS PhD FHEA MRCVS, is Lecturer in Veterinary 5
Education, Royal Veterinary College, The LIVE Centre, Hawkshead Lane, Hertfordshire, AL9 6
7TA; [email protected]; ORCID: 0000-0002-9395-8431. 7
Kate A Cobb, BVetMed PGCE MMedSci PhD MRCVS, is Lecturer in Teaching Learning and 8
Assessment, University of Nottingham School of Veterinary Medicine and Science, 9
Loughborough, Leics, LE12 5RA; [email protected]. 10
Liz H Mossop, BVM&S MMedSci (Clin Ed) PhD MAcadMEd MRCVS, is Associate Professor 11
of Veterinary Education, University of Nottingham School of Veterinary Medicine and Science, 12
Loughborough, Leics, LE12 5RA; [email protected]. 13
14
Key words: Clinical reasoning, simulation, veterinary education, veterinary medicine 15 Word count: 5223 16 17 18 19 20 21
22 23
Abstract 24
Clinical reasoning is an important skill for veterinary students to develop prior to graduation. 25
Simulation has been studied in medical education as a method for developing clinical 26
reasoning in students, but evidence supporting this is limited. This study involved the 27
creation of a contextualised standardised client simulation session aiming to improve the 28
clinical reasoning ability and confidence of final year veterinary students. Sixty-eight 29
participants completed three simulated primary-care consultations, with the client being 30
played by an actor and the pet by a healthy animal. Survey data showed 100% of participants 31
felt the session improved their ability to make clinical decisions. Quantitative clinical 32
reasoning self-assessment, performed using a validated rubric, triangulated this finding – 33
showing an improvement in student perception of several components of their clinical 34
reasoning skill level before and after the simulation. Blinded researcher analysis of the 35
consultation video-recordings found the ‘History-taking’ and ‘Making sense of data’ 36
(including differential diagnosis formation) components of the assessment rubric showed a 37
significant increase in ability. Thirty students took part in focus groups investigating their 38
experience within the simulation. Two themes arose from thematic analysis of this data: 39
Variety of reasoning methods and ‘It’sa different way of thinking’. The latter highlights 40
differences between the decision-making students practice during their time in education, and 41
the decision-making they will use once working in practice. The study findings suggest that 42
simulation can be used to develop clinical reasoning in veterinary students, and demonstrates 43
the need for further research in this area. 44
Introduction 46
The use of simulation in veterinary education has grown in the last 10 years. This has been 47
mainly driven by the increasing importance placed on communication training (1) and 48
clinical skills teaching, coupled with the overwhelming acceptance of the pedagogical value 49
of simulation within the fields of human medicine and nursing. It may also be due, in part, by 50
the increasing numbers of veterinary students at universities makes time practicing clinical 51
skills competitive and limited (2,3). However, simulation use within veterinary schools 52
remains very limited compared to other healthcare fields. 53
Simulated clients (SCs) are commonly used to develop communication skills in veterinary 54
students (1). Actors recreate the experience of conversing with a client so that students may 55
practice the techniques of history taking, dealing with conflict and breaking bad news. 56
Although effective at improving communication (4), SCs are rarely used for any other skill 57
development in veterinary education. 58
Clinical reasoning is the skill used when veterinary surgeons make a decision regarding the 59
diagnosis, treatment plan or prognosis of a patient (5). There are two cognitive processes a 60
practitioner can employ to solve these clinical problems – known as systems one and two 61
reasoning (6). System one is fast, unconscious and intuitive, whilst system two is slow, 62
logical and analytical (7). Whilst they can be used exclusively, they have been shown to be 63
most accurate when used in combination (8,9) – known as ‘dual processing’. As expertise in 64
clinical reasoning develops, students move away from a detailed pathophysiology-based view 65
of disease (system two), and begin to form readily-accessible illness scripts that permit a 66
form of diagnostic pattern-recognition (system one). However, experts retain the ability to to 67
switch back to a slower, logical method of decision making if they wish (dual-68
processing)(10,11). 69
Although thoroughly researched in medical domains, relatively little is known about 70
veterinary clinical reasoning (12,13). Even less is understood about how to ‘teach’ clinical 71
reasoning to veterinary students, thus most recommendations have been extrapolated from 72
medical research (14). This is not ideal, as it has been suggested that veterinary surgeons 73
integrate non-clinical factors such as finances and owner preferences to a greater degree than 74
their medical counterparts (12) – indicating different training needs. Vinten et al. (5) 75
conducted a qualitative investigation into clinical reasoning development at one UK 76
veterinary school – finding that graduates faced a steep learning curve when entering 77
practice. This is both supported (15) and refuted (16) by survey data from other authors. 78
Vinten et al. recommended included incorporating contextual factors into decision-making 79
training, and recreating the experience of responsibility for clinical outcomes – without which 80
students rely on clinicians present to prevent any harm to their patients. 81
Several studies have indicated that simulation might improve clinical reasoning in both 82
medical and nursing students (17–22). However, due to the inherent difficulties in 83
definitively measuring clinical reasoning, no research has provided strong enough evidenceto 84
be conclusive on this matter. There has been no research investigating the relationship 85
between simulation and clinical reasoning in veterinary students, to the authors’ knowledge. 86
This study aimed to assess the effect of novel primary care consultation simulation on the 87
clinical reasoning ability of final year veterinary students and explore the student experience 88
of clinical reasoning within a simulation scenario. Ethical approval was granted by the 89 University of Nottingham. 90 91 92 93
Methods 94
Simulation session design 95
The simulation was aimed at final year students, designed to recreate a first opinion small 96
animal consultation as closely as possible. The intended reasoning-based learning outcomes 97
were as follows: 98
1. Make clinical decisions confidently 99
2. Formulate differential diagnoses and diagnostic or treatment plans for a range of 100
clinical cases 101
3. Reflect on clinical decisions that have been made 102
Key features found to promote effective learning within simulations by Issenberg et al. (23) 103
were incorporated where possible (Table 1). Three cases were developed from genuine 104
patients examined and treated by one of the authors (CV) within a primary care veterinary 105
surgery. These were checked for authenticity by two experienced veterinary surgeons (a 106
summary of the cases is provided in table 2). Clients were played by trained actors and 107
patients played by healthy dogs belonging to the authors. 108
INSERT TABLES 1 AND 2 HERE 109
Prior to the simulation, students were provided with a very short description of each case (e.g. 110
‘weight loss’) in order to allow them to research the relevant topics, but no information or 111
tuition on clinical reasoning theory or methods. The session took place in a consultation room 112
within a small animal hospital, thus was already fully equipped. After an introduction and 113
familiarisation period, students were given a clinical history for their first patient, detailing 114
only previous treatment at the fictional surgery. When ready, the student collected the SC 115
and their pet from the hospital waiting room. The structure of the consultation was controlled 116
by the student and ended with the SC exiting the room. Each simulation lasted roughly 15 117
minutes. The students were instructed to treat the simulation as if it were a real consultation; 118
responding to the concerns of the client in an appropriate way, discussing possible diagnoses 119
and treatment options and prescribing any necessary medication. A 15-minute debriefing 120
using the model of Good Judgement (24) was then performed by a member of staff who had 121
observed the consultation through a live video feed. Each student participated in all three 122
cases in a randomised order. An overview of the simulation process for each student is shown 123
in Figure 1. 124
All students that undertook a placement at the small animal hospital during the 10-month 125
study period were required to take part in the simulation, but enrolling in the associated 126
research project was voluntary. Participants were separated into two cohorts – Group A took 127
part in the simulation within the first 6 months of their final year of study, group B within the 128
last 6 months. This was due to the timing of the study, which fell across two academic years, 129
but provided opportunity to observe the effect of the simulation at different points in the 130
curriculum. 131
Quantitative measurement of simulation impact 132
Due to the known difficulties objectively measuring clinical reasoning, three methods of data 133
collection were used in order to triangulate any findings. The Lasater clinical judgement 134
rubric (LCJR), developed by Lasater (25) was chosen to grade clinical reasoning ability 135
because a) it is specific for use within high fidelity simulation, allowing grading of physical 136
actions and conduct rather than written answers and b) it could be modified to give a 137
quantitative score of clinical reasoning skill. The components of the rubric were designed to 138
specifically relate to clinical reasoning – for example, the ‘History Taking’ component 139
measures directed questioning relevant to the case, rather than the associated communication 140
skills such as summarising or screening. 141
As the LCJR was developed to examine clinical judgement in human nursing, rather than 142
veterinary medicine, minor modifications were made to ensure it was suitable for a veterinary 143
application. These included changing of certain words (e.g. ‘patient’ to ‘client’) and the 144
removal of irrelevant areas of assessment (i.e. skills that would not be used). 145
The modified Lasater Clinical Judgement Rubric (mLCJR) and the three clinical cases (table 146
3) were piloted using a test simulation. One experienced veterinary surgeon was video-147
recorded completing the three simulated consultation cases. The rubric was then used to 148
assess the performance of the participant. No changes were necessary to the simulation cases 149
following the pilot study, but the mLCJR was modified to include a representative example 150
of student performance for each score category (appendix 1). 151
The mLCJR was used in two ways during the simulation. Firstly, students were asked to 152
score their own clinical reasoning ability pre and post simulation using the rubric. This was 153
performed immediately before the first consultation, and after the debriefing period of the last 154
consultation (self-assessment – SA). Secondly, the participant’s clinical reasoning was scored 155
by a researcher using the rubric (researcher assessment – RA). The first and third 156
consultations each student conducted were video recorded – a process the students were 157
familiar with from communication training earlier in the course. After completion of data 158
collection, these videos were blinded, randomised and scored by researcher CV; a small 159
animal veterinary surgeon experienced in teaching clinical reasoning. Ten percent of the 160
video recordings were also scored by a second researcher, also a veterinary surgeon, allowing 161
the interrater reliability to be calculated. This was done by calculating the Intraclass 162
Correlation Coefficient using SPSS statistics 22 (IBM). 163
To determine whether the data from groups A and B could be amalgamated, the difference 164
between the pre and post simulation scores of each student were calculated for both the SA 165
and RA. These were input into SPSS statistics 22 (IBM) and A Mann-Whitney U test 166
comparing the improvement of each group was performed on each mLCJR component 167
separately. There was a statistically significant difference in the score-change between the 168
two groups on the SA, so the data sets were not merged. There was not a significant 169
difference for the RA, so the data for groups A and B were combined. 170
The following methods were performed separately on groups A and B when evaluating the 171
SA and once on the combined data from both groups when analysing the RA. 172
The pre and post simulation scores were compared using a Wilcoxon Signed Ranks test. Each 173
component of the mLCJR was analysed individually. Median and mean averages were 174
calculated for each component, both pre and post simulation. 175
To determine whether the components could be summed to create an overall pre/post total for 176
each group, Cronbach’s alpha value of internal consistency was calculated. As all alpha-177
values fell above 0.7, the consistency was accepted within all four categories (Group A 178
SA/RA, Group B SA/RA) and the components summed (26). A Wilcoxon signed ranks test 179
was then performed on the totalled data. 180
Construct validation 181
To determine the construct validity of the mLCJR, a cohort of experienced veterinary 182
surgeons were tested using the rubric. A purposive sample of seven university staff members 183
that had over three years’ experience as a first-opinion small animal veterinary surgeon and 184
had worked in practice within the last 12 months were selected. All took part in one 185
simulated consultation and were video recorded. 186
The expert participants’ recordings were graded by a researcher (CV) and the median and 187
mean average total score calculated. Blinding was not possible as, due to the age of the 188
experts compared to the students, the identity of the staff was unavoidably clear. 189
To compare the expert and student performances, all student total scores were combined with 190
the expert total score data set. A Mann-Whitney U test was used to identify any significant 191
ability differences between the two groups. 192
Survey analysis of simulation impact 193
A Likert-scale survey was designed to collect student opinions about the simulation. Survey 194
responses were converted to numerical data for analysis, where Strongly disagree = 1 and 195
Strongly agree = 6. A Mann-Whitney U test indicated that group A and B should be further 196
analysed separately. 197
To determine if the questions could be summed to a total, Cronbach’s alpha was performed. 198
As both groups alpha values returned above 0.7 (26) the total score for each student was 199
calculated. For both cohorts, the median and mean averages were determined for each 200
question. The total percentage agreement with each question was then calculated. 201
INSERT FIGURE 1 HERE 202
Qualitative insights into simulation impact 203
Focus groups were conducted with 30 of the 68 students that took part in the simulation. 204
Participants were selected using convenience sampling, due to their busy schedule whilst on 205
final year work placements. Six focus groups were held, each with five participants. Each 206
focus group was held two days after the participants completed the simulation and was 207
optional. 208
The focus groups followed a semi-structured format and lasted between 30 to 60 minutes. 209
Questions focused on the experience of the students during the simulation; how the 210
experience differed from other experiences of decision-making during their training and how 211
participants felt they reasoned through the cases. All focus groups were audio recorded, 212
transferred electronically to a computer and then transcribed verbatim, by either an external 213
source or a researcher. Where transcription was done by an external source, the document 214
was checked by the researchers for accuracy. 215
The transcriptions for all focus groups were merged into one data set for thematic analysis. 216
Thematic analysis was performed using guidelines developed by Braun & Clarke (27). 217
Complete inductive code generation was performed by one researcher (CV), managed 218
through NVIVO (QSR, version 10). One focus group transcript was coded by a second 219
researcher (LM) and agreement reached in order to ensure consistent approach to coding. 220
Codes were then interpreted and grouped together by that researcher to form subthemes and 221
themes. These themes were iteratively revised and edited. Once complete, the themes were 222
reviewed by the remainder of the research group (KC, LM) and changes were made, which 223
prompted another round of iterative revision and editing. When finished, the group reviewed 224
the final themes once more and agreed on their interpretation. 225 226 227 228 229 230 231
Results 232
Sixty-eight students took part in the simulation – 32 in group A and 36 in group B. A 233
confidence interval of 95% was selected as a measure of statistical significance. 234
Student self-assessment 235
Group A reported significant improvement in all components of the mLCJR (Table 3). Group 236
B showed significant improvement in four out of eight components: History-taking, 237
Identifying abnormalities, Making sense of data and Well planned intervention (table 4). 238
INSERT TABLES 3 AND 4 HERE 239
Cronbach’s alpha showed acceptable reliability to sum total ‘before’ and total ‘after’ scores 240
(α=>0.7). A Wilcoxon Signed-Ranks Test indicated that post-simulation scores were 241
statistically significantly higher than pre-simulation scores for both groups (A: Z=-4.61, 242
p=<0.001; B: Z=-3.44, p=0.001). A Mann-Whitney test then showed that the level of 243
improvement was greater for group A (Mdn=1, Mn=1.88) than for Group B (Mdn=1, 244
Mn=1.26), U=340.0 p=.0006). 245
Researcher assessment 246
The two assessors reached an ICC of 0.894 (p=<0.05) after marking 10% of the video 247
recordings, indicating ‘almost perfect’ inter-rater reliability (28). 248
None of the mLCJR components showed a statistically significant difference in score 249
between groups A and B, so datasets were combined for further analysis. Within this 250
combined data, two mLCJR components showed significant improvement as a result of the 251
simulation: History taking and Making sense of data (table 5). 252
INSERT TABLE 5 HERE 253
Cronbach’s alpha showed acceptable reliability to sum total ‘before’ (α=67) and total ‘after’ 254
scores (α=0.75). The Wilcoxon signed ranks test showed no significant difference between 255
total scores (table 5). 256
Construct validation of mLCJR 257
Seven expert participants took part in the validation simulation. A Mann-Whitney test 258
indicated that total scores were higher for the expert group (Mdn=31.00) than the student 259
group (Mdn=27.00), U=43.00 P=0.003. This suggests the mLCJR has an acceptable construct 260
validity. 261
Survey 262
A Mann-Whitney test showed the two groups answered nine questions significantly 263
differently, therefore data was not merged (table 6). In both groups, 100% of students 264
reported feeling more confident in making decisions, reaching a diagnosis and forming a 265
treatment plan. The median and mean averages show group A answered all questions with a 266
higher level of agreement than group B (table 6). 267
INSERT TABLE 6 HERE 268
Cronbach’s alpha showed excellent internal consistency of both group A (α=0.84) and group 269
B (α=0.86). A Mann-Whitney test indicated that the total level of agreement (table 6) was 270
greater for group A (Mdn=82.00) than for group B (Mdn=77.00), U=284.50 P=0.001. 271
Qualitative data 272
Two key themes emerged from the focus group data, which are described below with 273
supporting quotes from the transcripts. Each focus group has been assigned a code - FG1, 274
FG2, FG3, FG4, FG5 or FG6. 275
Theme one: ‘It’s a different way of thinking…’ 276
During the analysis, it became clear that the clinical reasoning taking place within the 277
simulation had many differences from other decision-making experiences students had had in 278
the curriculum. Three key factors were described as being novel. Firstly, the students 279
described the simulation as being their first experience of making clinical decisions alone. 280
They spoke of using the clinician usually present in consultations as a ‘safety-net’; ensuring 281
that any mistakes they make are corrected before they have consequences. Thus, they felt 282
their decisions were always ‘checked’ and approved. 283
‘(In other consultations) you have got that safety net behind you… if you say 'I'm thinking
284
about this' and they say 'Well maybe, but think about...' you have always got someone there
285
pointing you in the right direction.’ FG1
286
‘(In the simulation) all the responsibility is on you – it’s the first time we have properly had it
287
all on us in a way… because you have always got a clinician as a back-up in every other case
288
we’ve been doing.’ FG2
289
Students felt that having a clinician present in other consultations has removed their sense of 290
case responsibility. Being alone in the simulation helped to create the experience of having 291
sole charge over decision-making – despite the fact the clients and patients were not real. 292
‘I just found it quite generally daunting taking on the consult and prime responsibility…where
293
you did not have anyone to rely upon for the first time.’ FG6
294
Secondly, the students were not used to making clinical decisions in pressurised situations. 295
They felt that having a client in the consultation room forced them to make decisions faster. 296
Students described the consultations they perform with clinicians (which are normally given 297
triple the standard appointment time allowance) as slow-paced, and thus the skill of thinking 298
under pressure is not practiced. 299
‘You have to make quite a quick decision (in the simulation)... Where I think with (clinicians)
300
you can have a nice chat and discuss your different options and then decide which ones are
301
sensible to go with.’ FG1
302
‘(In the simulation) you have got to make the decision there and then, you haven’t got time to
303
go away and think about it…’ FG3
304
Students also commented that the pressure of the consultation did not allow for the same 305
reasoning processes they have developed on paper through case based learning and 306
assessment. It was suggested that thinking ‘in your head’ is harder than reasoning on paper or 307
similar, and thus the opportunity to practice it was valuable. 308
‘It’s a different way of thinking though, isn’t it, because when you’ve, when you write it on
309
paper you’re working through in stages, whereas if you’re in conversation you have to skip
310
half of that stuff’ FG5
311
‘It is one thing being able to write on a piece of paper what you are thinking and sit there and
312
look at what you have put down, but it is another thing processing it all in your brain and
313
your head and thinking about what you need to ask and then thinking of what other possible
314
things it could be.’ FG6
315
The integration of situational factors was the third aspect of clinical reasoning within the 316
simulation that students found novel. This involved combining their decision-making skills 317
with communication, considering the owners needs and administrative tasks. 318
‘You are multi-tasking in the simulation because you are also thinking what am I projecting
319
to the client? How am I going to explain it to the client? Am I being clear?’ FG1
320
‘We’ve never, ever had to deal with money before, we’ve never had to think about prices, or
321
trade names…’ FG4
322
‘On paper you could be like ‘Go home on a bland diet, whatever’ – but (in the simulation)
323
there is a client, waiting, stood there, probably expecting antibiotics or something… so that’s
324
different because you have to manage client expectations.’ FG3
325
Students appear to process information differently to draw conclusions within the simulation 326
compared to case-based learning sessions, examinations and clerkship consultations. They are 327
learning to think in different way to cope with the time pressures and multi-tasking required. 328
Theme two: Variety of reasoning methods 329
Students reported using both system one and system two reasoning. They were not 330
consciously aware of the difference, but it became clear through their discussion that this was 331
the case. 332
‘Sometimes I find it hard to explain how I came up with the solution, sometimes it does just
333
ping there like ‘Oh I think this is what I should do’.’ FG3
334
‘My brain doesn’t just go like (clicks fingers) … it always takes me a longer time for some
335
reason.’ FG4
336
It was also clear from the data that there was a degree of case specificity affecting the ability 337
to make clinical decisions. Students disagreed on which case was most complicated, and their 338
opinions generally reflected their level of knowledge about each pathology. 339
‘I felt that the one consult that I did better in was the one that I knew more about and you felt
340
more comfortable with.’ FG5
341
342 343
Discussion 344
The effect of standardised client simulation on clinical reasoning development 345
The RA showed improvement in only two of the components of the mLCJR – history taking 346
and making sense of data. The latter of these focuses on the formation of differential 347
diagnoses, arguably a key aspect of clinical reasoning and one the session aimed to improve. 348
The former, history taking, is a skill that the fifth year students involved in the simulation 349
were already expected to be proficient in. One explanation for the noticeable improvement in 350
history taking may be that the task actually required the formation of differential diagnosis in 351
order to ask the necessary question to rule each in/out. Although students have practiced the 352
communicatory tasks of history taking previously, they had limited opportunities to combine 353
this with a diagnostic task. This theory is supported by the work of Nendez et al. (29). They 354
found that the diagnostic accuracy of students, residents and practitioners all decreased when 355
only a ‘chief complaint’ was provided and further data collection was required, opposed to a 356
full clinical vignette. The authors discovered the reason behind poor performance with chief 357
complaint scenarios was the failure to gather sufficient information during the history taking 358
process, despite being given (when asked) more information than the vignettes provided. The 359
authors conclude that the teaching of history taking should be integrated with reasoning tasks, 360
so that students practice using the two in conjunction and thus are able to apply this model 361
when in practice. If extrapolated to veterinary medicine, this theory could explain the 362
improvement in history taking, despite it not being a focus of the simulation; i.e. by 363
reviewing the formation of differential diagnoses during the debriefing, the ability to 364
structure data gathering also improved. In an investigation of the structure of veterinary 365
consultations, Everitt (12) found that the history taking process was interweaved with the 366
physical examination – suggesting that the former is used to inform the latter and vice versa. 367
This theory further supports the increase in history taking ability being an indicator of clinical 368
reasoning improvement. 369
The SA and RA do not appear to agree on the level of development during the simulation. 370
One possibility is that students have over-estimated their improvement, or simply gained 371
confidence but not measurable skill. A second possibility is that case specificity affected the 372
student’s objective skill level between cases. Case specificity was first noted by Elstein et al. 373
(30) when they observed that the diagnostic ability of a physician varied – scoring well on 374
one case examination was not an indicator of future performance. The implication of this was 375
that knowledge plays a role in clinical reasoning; it is not simply a generalizable skill (31). 376
Further research has shown that actually a combination of knowledge and general problem-377
solving ability is needed for successful reasoning (31–33), however no studies exclude the 378
need for domain specific knowledge. If this theory were applied to this study, a student that 379
had greater knowledge about, for example, idiopathic epilepsy would be more likely to 380
perform well during that case simulation, regardless whether it was their first or last 381
consultation. If their knowledge of acute diarrhoea and weight loss causes was significantly 382
lower, any reasoning skill development might become negligible. As the students in this 383
study were provided with a case-list several days prior to the simulation it was expected they 384
would research the topics, thus reducing the effect of subject-specific knowledge. However, 385
whether or not the students did partake in revision was not measured and so it is difficult to 386
estimate the influence of case specificity. If this study were to be repeated, providing reading 387
material or a lecture on the topics to be addressed in the forthcoming consultations and then a 388
test of mastery would help to reduce case specificity. There would always be, however, the 389
effect of personal experience on knowledge and decision-making that would case some 390
degree of bias. 391
One further factor may have contributed to the difference between the RA and SA score 392
improvement. Three components of the mLCJR – Examination, Identifying abnormalities and 393
Prioritising data – had a RA ‘first consultation’ median score of four; the highest possible 394
mark. This means that it was not possible for students to improve in those areas (in a way that 395
was recognisable on the mLCJR). It is likely that this arose due to a mismatch between 396
student ability and simulated consultation difficulty. In future work, increasing the difficulty 397
of the cases could reduce this effect. As it may not be possible to manipulate the physical 398
examination task, this component might need to be removed from the mLCJR. 399
Differences between the research groups 400
Group A (early) reported a significantly larger degree of improvement than group B (late) 401
during the SA within four categories: History taking, examination, calm confident manner 402
and clear explanation. It can be argued that these four components of the mLCJR are covered 403
well within veterinary curricula – particularly in the final year of the course, when students 404
engage in workplace-based learning. The fact that group B did not improve as much in these 405
four categories as group A suggests that teaching and repetitive practice in fifth year might 406
improve their perceived ability in these areas to a level at which they felt proficient by the 407
time the simulation was conducted. The remaining components - identifying abnormalities, 408
prioritising data, making sense of data and forming a well-planned intervention – represent 409
key mental tasks during clinical decision-making. These components showed the same 410
increase within both group A and B, implying that there is little perceived improvement in 411
ability during fifth year. Overall, this suggests that some components of clinical reasoning are 412
being developed by the workplace-based learning, but essential mental processes are 413
remaining unchanged throughout. This difference is not mirrored in the RA results, in which 414
both groups of students performed equally. 415
Qualitative data results 416
Within the theme ‘It’s a different way of thinking…’. Students claimed their clinical 417
reasoning process was different within a simulation, compared to consultations with 418
clinicians, case-based learning or examinations. This has important consequences for how 419
clinical reasoning should be taught, as the simulation closely resembles the day-to-day work 420
of a veterinary surgeon and thus the way clinical reasoning will be used frequently upon 421
graduation. 422
Students described the pressure of making decisions quickly within the simulation as 423
something new that they have not experienced elsewhere; a way of reasoning that required 424
different thought processes than they were used to. It is known that stress affects human 425
decision-making – increasing the amount of risk-taking behaviour observed (34). In these 426
circumstances, subjects use heuristics more frequently (35), possibly due to working memory 427
overload. Studies of both veterinary surgeons and human physicians have shown that they 428
suffer greater levels of stress than the general population, especially those recently graduated 429
(36–38). The combination of these two factors – high stress and the impact of stress on 430
decision-making – suggests educators need to be giving students opportunity to practice 431
clinical reasoning under pressure. If the process of reasoning is different when time is not 432
limited, then efforts to develop clinical reasoning in relaxed settings will not prepare students 433
for making decisions in the real world. Simulation is known for causing stress in students – 434
generally perceived as a negative consequence (20,39). However, this ‘side-effect’ of 435
simulation-based education could be utilized for the students benefit. The timing of such an 436
intervention would be critical – subjecting a student to decision-making under pressure before 437
they are capable would only damage their confidence. But, for a student already competent at 438
clinical reasoning in the classroom and clinic, simulation may provide the last key situation in 439
which to master their skill. 440
Another major finding of this theme is that the simulation experience was the first time 441
students had felt fully responsible for their own clinical decisions. Even when they are given 442
opportunities to make decisions within WBL consultations, the students report a sense of 443
security from the clinician present that prevents them from emotionally investing in their 444
decision. The same problem has been reported previously in medicine, where the ‘simplistic’ 445
approach to teaching clinical reasoning generates a ‘sterile academic environment which 446
avoids feelings of responsibility for any morbidity or mortality experienced by the patient as 447
a consequence of making an inappropriate diagnosis’ (40). Again, the effect of diminished 448
responsibility is that students practice a cosseted form of clinical reasoning that is not fully 449
representative of the skill they will need to use in practice. Thus, when they graduate, they 450
are underprepared. 451
Student participants found situated decision-making another new challenge. They found 452
incorporating owner factors particularly novel, alongside the need for multi-tasking. This 453
probably results from the isolated nature of other clinical reasoning experiences - normally 454
students make clinical decisions in an artificial environment where their only task is to 455
develop an appropriate case management plan. This allows them to focus all their 456
concentration on the decision-making process, which is not often possible in reality. On top 457
of this, students do not always have the opportunity to complete clinical notes, prescribe and 458
dispense drugs or calculate costs when participating in real consultations during clerkships. 459
These form ‘distractors’ that interfere with clinical reasoning, however students rarely 460
practice incorporating them into decision-making. Several studies have shown that contextual 461
factors impact clinical decision-making (41–43), meaning teaching students to recognize and 462
respond to these distractors is important. Again, students cited the SC simulation as an 463
effective way to develop multi-tasking ability. 464
The theme ‘variety of reasoning methods’ developed from discussing with the students how 465
they made decisions within the simulation. There were various methods described, including 466
both systems one and two. This is not surprising, as Coderre et al. (44) not only showed that 467
both system one and system two methods were used by students, but also that diagnostic 468
accuracy was significantly higher when using the former. A later study by Ark et al. (8) found 469
that students using dual process reasoning were most diagnostically successful. This has 470
implication for veterinary education, as it indicated that system one reasoning should not be 471
discouraged; in fact, students should be aware of it so they may utilise it correctly. 472
473
Limitations 474
To date, there is no published method of measuring clinical reasoning ability that has an 475
acceptable construct validity. This study aimed to increase the validity and reliability of the 476
results by using four methods of data collection to triangulate results and by attempting to 477
evaluate the construct validity of the rubric used. However, this remains the biggest limitation 478
of this study and the results must be interpreted accordingly. 479
480
The limitations of using self-reported data also need to be considered. These relate to the 481
confines of introspection, and the ability to understand one’s own subconscious decision-482
making process. Again, the use of triangulation minimizes the effect of any inaccuracy, but 483
does not eliminate it. The fact that the focus group facilitator also facilitated the simulation 484
may have impacted on the responses of participants. This possibly deterred students from 485
criticising the session, however, all students were encouraged to reflect on the experience 486
honestly and were made aware that their data would be anonymised and only used for the 487
purpose of this research project. Finally, due to the time-span of this study, peer disclosure of 488
the simulation structure could not be prevented using quarantine methods. The impact of this 489
was minimised using two strategies: 1) students were asked not to discuss the cases outside of 490
the simulation and 2) all participants were given information about the consultation topics 491
before the simulation, thus reducing the impact of knowledge differences on the scores 492 achieved. 493 494 Conclusion 495
In summary, this study has shown that standardised client simulation can be used to increase 496
student confidence in clinical reasoning ability. There is also some evidence that simulation 497
objectively improves some aspects of clinical reasoning, including differential diagnosis 498
formation. This study also highlights the differences between the decision-making students 499
practice during their time in education, and the decision-making they will use once working 500
in practice. High fidelity simulation is indicated as one successful way to align the curriculum 501
content to the career needs 502
503
504 505 506
References 507
1. Gray C, Blaxter A, Johnston P, Latham C, May S, Phillips C, et al. Communication 508
Education in Veterinary Education in the United Kingdom and Ireland: The NUVACS 509
Project Coupled to Progressive Individual School Endeavors. J Vet Med Educ. 2006 510
Mar;33(1):85–92. 511
2. Kneebone R, Baillie S. Contextualized simulation and procedural skills: a view from 512
medical education. J Vet Med Educ. 2008;35(4):595–8. 513
3. Byron J, Johnson S, Allen C, Brilmyer C, Griffiths R. Development and pilot of Case 514
Manager: a virtual-patient experience for veterinary students. J Vet Med Educ. 515
2014;41(3):225–32. 516
4. Latham C, Morris A. Effects of formal training in communication skills on the ability 517
of veterinary students to communicate with clients. Vet Rec. 2007 Feb 10;160(6):181– 518
6. 519
5. Vinten CEK, Cobb KA, Freeman SL, Mossop LH. An investigation into the clinical 520
reasoning development of veterinary students. J Vet Med Educ. 2016;43(4). 521
6. Croskerry P, Petrie D, Reilly J, Tait G. Deciding about fast and slow decisions. Acad 522
Med. 2014;89(2):197–200. 523
7. Canfield P, Whitehead M, Johnson R, OBrien C, Malik R. Case-based clinical 524
reasoning in feline medicine: 1: Intuitive and analytical systems. J Feline Med Surg. 525
2016;18(1):35–45. 526
8. Ark T, Brooks L, Eva K. Giving learners the best of both worlds: Do clinical teachers 527
need to guard against teaching pattern recognition to novices? Acad Med. 528
2006;81(4):405–9. 529
9. Ark T, Brooks L, Eva KW. The benefits of flexibility: The pedagogical value of 530
instructions to adopt multifaceted diagnostic reasoning strategies. Med Educ. 531
2007;41(3):281–7. 532
10. Schmidt H, Norman G, Boshuizen H. A cognitive perspective on medical expertise: 533
theory and implication. Acad Med. 1990;65(10):611–21. 534
11. Schmidt H, Rikers R. How expertise develops in medicine: knowledge encapsulation 535
and illness script formation. Med Educ. 2007;41(12):1133–9. 536
12. Everitt S. Clinical decision making in veterinary practice. University of Nottingham; 537
2011. 538
13. Vandeweerd J, Vandeweerd S, Gustin C, Keesemaecker G, Cambier C, Clegg P, et al. 539
Understanding Veterinary Practitioners’ Decision-Making Process: Implications for 540
Veterinary Medical Education. J Vet Med Educ. 2012;39(2):142–51. 541
14. May S. Clinical Reasoning and Case-Based Decision Making: The Fundamental 542
Challenge to Veterinary Educators. J Vet Med Educ. 2013;40(3):200–9. 543
15. Cobb K, Brown G, Hammond R, Mossop L. Alumni-based evaluation of a novel 544
veterinary curriculum: are Nottingham graduates prepared for clinical practice? Vet 545
Rec Open. 2015;2(2):1–9. 546
16. Gilling M, Parkinson T. The transition from veterinary student to practitioner: A 547
“make or break” period. J Vet Med Educ. 2009;36(2):209–15. 548
17. Cant R, Cooper S. Simulation-based learning in nurse education: systematic review. J 549
Adv Nurs. 2010;66(1):3–15. 550
18. Steadman R, Coates W, Huang Y, Matevosian R, Larmon B, McCullough L, et al. 551
Simulation-based training is superior to problem-based learning for the acquisition of 552
critical assessment and management skills. Crit Care Med. 2006;34(1):151–7. 553
19. Powell-Laney S, Keen C, Hall K. The Use of Human Patient Simulators to Enhance 554
Clinical Decision-making of Nursing Students. Educ Heal. 2012;25(1):11–5. 555
20. Lasater K. High-fidelity simulation and the development of clinical judgment: 556
Students’ experiences. J Nurs Educ. 2007;46(6):269–76. 557
21. Gibbs J, Trotta D, Overbeck A. Human patient simulation versus case study: Which 558
teaching strategy is more effective in teaching nursing care for the hypoglycemic 559
patient? Teach Learn Nurs. 2014;9(2):59–63. 560
22. Yuan H, Williams B, Man C. Nursing students’ clinical judgment in high-fidelity 561
simulation based learning: A quasi-experimental study. J Nurs Educ Pract. 562
2014;4(5):7–15. 563
23. Issenberg B, McGaghie W, Petrusa E, Lee Gordon D, Scalese R. Features and uses of 564
high-fidelity medical simulations that lead to effective learning: a BEME systematic 565
review. Med Teach. 2005;27(1):10–28. 566
24. Rudolph J, Simon R, Rivard P, Dufresne R, Raemer D. Debriefing with good 567
judgment: combining rigorous feedback with genuine inquiry. Anesthesiol Clin. 2007 568
Jun;25(2):361–76. 569
25. Lasater K. Clinical judgment development: using simulation to create an assessment 570
rubric. J Nurs Educ. 2007;46(11):496–503. 571
26. Bland J, Altman D. Cronbach’s alpha. Vol. 314, British medical journal (Clinical 572
research ed.). 1997. p. 572. 573
27. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 574
2006;3(2):77–101. 575
28. Royal K, Hecker K. Understanding Reliability: A Review for Veterinary Educators. J 576
Vet Med Educ. 2015 Nov 11;e(e):1–4. 577
29. Nendaz M, Raetzo M, Junod A, Vu N. Teaching Diagnostic Skills: Clinical Vignettes 578
or Chief Complaints? Adv Heal Sci Educ. 2000;5(1):3–10. 579
30. Elstein A, Shulman L, Sprafka S. Medical problem solving: An analysis of clinical 580
reasoning. Vol 2. Cambridge, MA: Harvard University Press; 1978. 581
31. Wimmers P, Fung C. The impact of case specificity and generalisable skills on clinical 582
performance: a correlated traits−correlated methods approach. Med Educ. 2008 583
Jun;42(6):580–8. 584
32. Norman G, Bordage G, Page G, Keane D. How specific is case specificity? Med Educ. 585
2006 Jun 23;40(7):618–23. 586
33. Dory V, Gagnon R, Charlin B. Is case-specificity content-specificity? An analysis of 587
data from extended-matching questions. Adv Heal Sci Educ. 2010;15(1):55–63. 588
34. Starcke K, Brand M. Decision making under stress: A selective review. Neurosci 589
Biobehav Rev. 2012 Apr;36(4):1228–48. 590
35. Keinan G. Decision making under stress: Scanning of alternatives under controllable 591
and uncontrollable threats. J Pers Soc Psychol. 1987;52(3):639–44. 592
36. Bartram D, Yadegarfar G, Baldwin D. Psychosocial working conditions and work-593
related stressors among UK veterinary surgeons. Occup Med (Chic Ill). 594
2009;59(5):334–41. 595
37. Gardner D, Hini D. Work-related stress in the veterinary profession in New Zealand. N 596
Z Vet J. 2006;54(3):119–24. 597
38. Wall TD, Bolden R, Borrill S, Carter A, Golya D, Hardy G, et al. Minor psychiatric 598
disorder in NHS Trust staff::Occupational and gender differences. Br J Psychiatry. 599
1997;171(December):519–23. 600
39. DeCarlo D, Collingridge D, Grant C, Ventre K. Factors Influencing Nurses’ Attitudes 601
Toward Simulation-Based Education. Simul Healthc J Soc Simul Healthc. 602
2008;3(2):90–6. 603
40. Patel R, Sandars J, Carr S. Clinical diagnostic decision-making in real life contexts: A 604
trans-theoretical approach for teaching: AMEE Guide No. 95. Med Teach. 605
2014;37(3):211–77. 606
41. Durning S, Artino A, Boulet J, Dorrance K, van der Vleuten C, Schuwirth L. The 607
impact of selected contextual factors on experts’ clinical reasoning performance (does 608
context impact clinical reasoning performance in experts?). Adv Heal Sci Educ. 609
2012;17(1):65–79. 610
42. Sibbald M, Panisko D, Cavalcanti R. Role of clinical context in residents’ physical 611
examination diagnostic accuracy. Med Educ. 2011;45(4):415–21. 612
43. Durning S, Artino A, Pangaro L, van der Vleuten C, Schuwirth L. Context and clinical 613
reasoning: Understanding the perspective of the expert’s voice. Med Educ. 614
2011;45(9):927–38. 615
44. Coderre S, Mandin H, Harasym P, Fick G. Diagnostic reasoning strategies and 616
diagnostic success. Med Educ. 2003 Aug;37(8):695–703. 617
618 619
620
Feature Description Implications for the design
of this study
Feedback Providing a form of
feedback to the learners regarding their performance
Detailed personal feedback was given to each student after every simulated consultation by the facilitator
Repetitive practice Multiple opportunities for students to practice tasks - must be with the aim of improvement
Each student took part in three simulated consultations to allow them to practice decision-making multiple times and implement feedback given Capture clinical variation Portraying a variety of
clinical cases to maximise case exposure
The three consultations the student took part in all simulated different clinical cases
Controlled environment An environment where mistakes can be made safely and the facilitator can focus on the student, not the patient
The simulation was completely controlled – errors could be made without patient consequences Individualised learning Students should be active
participants in a simulation experience that is
individualised to each students needs
For this reason, students took part in the simulation alone and did not passively observe other students completing the simulation Simulator validity The simulation must have a
high fidelity and be comparable to a genuine experience
The simulation was designed to be as high fidelity as possible – including the absence of peers/facilitators in the consultation area Table 1: Components identified by Issenberg et al. (2005) to promote effective learning 621
during simulation that were incorporated into the simulated consultation design. 622
623 624
625
Table 2: Summary of the three cases developed for the standardised client simulation. 626
627 628
Case Signalment and
history Most likely diagnosis Appropriate treatment plan example Owner considerations Acute
diarrhoea 5-year-old dog Watery diarrhoea lasting two days No other relevant history Clinical exam normal Dietary indiscretion
Advise the owner to feed a bland diet (e.g. chicken breast) and administer
digestive support paste (e.g. Protexin Pro-kolin) twice daily according to weight
Usually seen by the senior vet, who always prescribes antibiotics for diarrhoea.
Seizure 3-year-old dog First seizure yesterday No other clinical signs, change in behaviour or relevant history Clinical exam normal Idiopathic epilepsy
Offer the owner a blood test
(biochemistry, haematology minimum) and advise monitoring at home for further seizure activity Has no insurance and can spend a maximum of £75 during this visit Weight loss and polydipsia 9-year-old dog 6 month history of slow but progressive weight loss No historical cause for weight loss Observed drinking
more water than usual latterly Clinical exam normal Diabetes mellitus/ Chronic kidney disease
Advise the owner to submit a urine sample for dipstick/specific gravity testing and recommend a blood test (biochemistry and haematology minimum) Mother recently died from cancer so is extremely sensitive to the possibility of tumours
629
Figure 1: The overall simulation session process; repeated for each student 630 631 632
Consent form
Self-assessment 1
Briefing
Scenario 1
Debriefing
Scenario 2
Debriefing
Scenario 3
Debriefing
Self-assessment 2
Survey
Researcher
assessment
(at a later date)
Table 3: Group A pre and post simulation self-assessment scores, with results of the 633
Wilcoxon signed-ranks test to determine if the difference between pre/post-simulation self-634
assessment scores is statistically significant. *P-value shows a statistically signficant 635 differnece (≤0.05) 636 637 638 639 mLCJR Component Median (mean) pre-sim score Median (mean) post-sim score Wilcoxon signed-ranks test statistic (Z score) P-value History taking 2.00 (2.47) 3.00 (3.13) -4.36 <0.001* Examination 2.00 (2.16) 3.00 (2.81) -4.38 <0.001* Identifying abnormalities 2.00 (2.03) 3.00 (2.75) -4.23 <0.001* Prioritising data 2.50 (2.53) 3.00 (2.78) -2.14 0.033* Making sense of data 2.00 (2.38) 3.00 (2.72) -2.40 0.016* Well planned intervention 2.00 (2.19) 3.00 (2.78) -3.34 0.001* Calm, confident manner 3.00 (2.59) 3.00 (3.03) -3.13 0.002* Clear explanations 3.00 (2.75) 3.00 (3.22) -4.61 <0.001* Total 20.50 (21.53) 25.00 (25.91) -4.61 <0.001*
Table 4: Group B pre and post simulation self-assessment scores, with results of the 640
Wilcoxon signed-ranks test to determine if the difference between pre/post-simulation self-641
assessment scores is statistically significant. *P-value shows a statistically signficant 642 differnece (≤0.05) 643 644 645 mLCJR
Component Median (mean) pre-sim score (mean) post-Median sim score Wilcoxon signed-ranks test statistic (Z score) P-value History taking 3.00 (2.83) 3.00 (3.11) -2.50 <0.012* Examination 3.00 (2.60) 3.00 (2.77) -1.90 0.057 Identifying abnormalities 2.00 (2.31) 3.00 (2.74) -3.27 <0.001* Prioritising data 3.00 (2.63) 3.00 (2.77) -1.51 0.131 Making sense of data 3.00 (2.49) 3.00 (2.71) -2.14 0.032* Well planned intervention 2.00 (2.23) 3.00 (2.69) -3.77 <0.001* Calm, confident manner 3.00 (2.89) 3.00 (3.97) -1.00 0.317 Clear explanations 3.00 (2.97) 3.00 (3.11) -1.67 0.095 Total 23.00 (23.57) 26.00 (25.66) -3.44 0.001*
mLCJR Component First consultation median (mean) score Third consultation median (mean) score Wilcoxon signed ranks test statistic (Z score) P-value History taking 2.00 (2.50) 3.00 (2.93) -3.00 0.003* Examination 4.00 (3.50) 4.00 (3.51) -0.01 0.992 Identifying abnormalities 4.00 (3.75) 4.00 (3.55) -1.57 0.116 Prioritising data 4.00 (3.60) 4.00 (3.61) -0.12 0.906 Making sense of data 2.00 (2.75) 3.50 (3.13) -2.16 0.031* Well planned intervention 3.00 (2.85) 2.00 (2.84) -0.49 0.625 Calm, confident manner 3.00 (3.13) 3.00 (3.09) -0.41 0.684 Clear explanations 3.50 (3.38) 3.00 (3.18) -1.90 0.058 Total 25.50 (25.47) 26.00 (25.84) -0.50 0.619
Table 5: First/third simulated consultation scores according to the researcher-assessment, 646
with results of the Wilcoxon signed ranks test to determine if the difference between 647
first/third consultation researcher-assessment scores is statistically significant (P≤0.05). *P-648
value shows a statistically signficant difference (≤0.05) 649
650 651
Question Group A n=32 Group B n=35 Mann-Whitney test statistic P-value median (mean) score Percentage
agreement median (mean) score Percentage agreement The session was enjoyable 6.00 (5.53) 100.00 5.00 (5.17) 100.00 378.00 0.010* The session
was a good use of my time 6.00 (5.72) 100.00 (5.54) 6.00 100.00 463.50 0.144 I would like to participate in a session like this again 6.00 (5.69) 100.00 5.00 (5.17) 97.10 323.00 0.001* My knowledge improved during the session 6.00 (5.53) 100.00 5.00 (4.97) 94.30 333.50 0.002* My practical skills improved during the session 5.00 (4.72) 96.90 4.00 (4.29) 88.60 411.00 0.043* My overall confidence in making decisions improved during the session 5.50 (5.41) 100.00 5.00 (5.03) 100.00 392.50 0.021* My overall ability to reach a diagnosis has improved as a result of the session 5.00 (5.09) 100.00 5.00 (4.77) 100.00 420.00 0.049* My overall ability to form a treatment plan has improved as a 5.00 (5.00) 100.00 5.00 (4.83) 100.00 491.00 0.341
result of the session I feel more prepared to undertake small animal consultations now 5.50 (5.41) 100.00 5.00 (5.20) 100.00 460.00 0.164 I found the session challenging 5.00 (5.03) 96.90 5.00 (4.61) 97.10 415.00 0.051 I found the session demoralising 1.00 (1.42) 0.00 2.00 (1.74) 0.00 405.00 0.030* I found the session and scenarios unrealistic 1.00 (1.44) 6.20 2.00 (1.65) 2.90 429.00 0.060 I felt embarrassed participating in the session 1.00 (1.78) 15.60 2.00 (2.14) 20.00 435.50 0.092 The feedback sessions were informative 6.00 (5.87) 100.00 5.00 (5.31) 97.10 276.00 <0.001* The feedback sessions were demoralising 1.00 (1.06) 0.00 (1.06) 1.00 0.00 338.00 <0.001*
Table 6: Median and mean average ratings for each survey question, percentage agreement 652
with each questions and results of Mann-Whiney test to determine if groups A and B 653
answered the survey differently. *P-value shows a statistically signficant difference (≤0.05) 654 655 656 657 658 659 660
Appendix 1 661 Component Score 1 2 3 4 History taking Is ineffective at taking a history. Obtains very limited information from the owner.
E.g. Only asks one or two of the mark sheet history questions.
Asks SOME required questions, but misses a few important ones out. Seems unsure what information to ask for and may ask irrelevant
questions.
e.g. Does not ask about water intake when faced with the weight loss case
Asks MOST required questions, but
occasionally does not follow up or clarify important leads. May miss one minor point, but asks all vital questions.
e.g. Does not ask about in-contact animals when faced with the D+ case
Asks ALL relevant questions when taking a history.
e.g. Asks all questions on the mark sheet
Examination
Examination is very limited, only one or two components are checked.
e.g. Only auscultates chest
Performs a LIMITED clinical examination. Important aspects of the exam are missed out.
e.g. Does not perform any neurological examination when faced with the seizing case
Performs a THOROUGH clinical examination; a few minor components are missed.
e.g. Does not check lymph nodes on any case
Performs a COMPLETE clinical examination, does not miss any components relevant to the case.
e.g. Completes all points on the mark scheme
Identifying abnormalities
Misses the importance of clinical findings – unjustly dismisses them.
e.g. Not appreciating significant weight loss that requires investigation in the weight loss case
Recognises SOME abnormalities, but
overlooks some important findings from the
history/exam.
e.g. Not noting polydipsia when faced with the weight loss case
Recognises MOST abnormalities that need to be considered, missing only minor aspects.
e.g. Not noting lethargy in the diarrhoea case
Recognises ALL problems that need to be addressed.
e.g. Identifies all relevant abnormalities
Prioritising data
Does not know which findings to concentrate on, prioritises an unimportant problem over the relevant issue – may not attend to the main problem.
e.g. Focusing on lack of flea treatment at length during the weight loss case
Attempts to focus on the main problem, but gets distracted. Alternatively, does not prioritise relevant findings as important.
e.g. Does not prioritise polydipsia as a problem when discovered in history of the weight loss case
Generally concentrates on the most important
findings, but does talk about irrelevant aspects of the exam/history
BRIEFLY.
e.g. Recommending worming when faced with the acute D+ case (except as general
recommendation to worm regularly)
Just discusses and forms a treatment plan for the relevant findings.
e.g. Only discusses aspects directly related to the current problem
Making sense of data
Struggles to interpret history and exam findings. Is unsure how to proceed. Does not determine a feasible way to proceed with the case.
e.g. Sends owner of weight loss case home with view to monitor weight over coming months
Attempts to interpret the clinical findings, but misses an IMPORTANT differential diagnoses or includes irrelevant ones.
e.g. Does not consider toxin ingestion when facing seizing case
Is able to interpret the history and clinical exam to form several
differential diagnoses, but may miss a MINOR differential or include a differential that is very low in likelihood.
e.g. Considers worm infestation a differential for acute D+
Is able to interpret the history and clinical exam to form a set of accurate differential diagnoses.
e.g. Clearly has
considered all relevant differential diagnoses when deciding how to proceed with case
Well planned intervention
Treatment plan is not acceptable treatment for the case.
e.g. Prescribing
antibiotics when facing acute D+ case
Treatment/investigation is not the most appropriate for the case, but some aspects are correct and will aid
diagnosis/treatment.
e.g. Not conducting urinalysis on patient with
Treatment/investigation plan is correct for the case, but there may be minor, aspects missed or incorrectly included.
e.g. Not advising Prokolin for acute D+ case
Treatment choice ideal for case (considering animal and owner factors).
e.g. Follows treatment plan on mark sheet
Appendix 1: The modified Lasater Clinical Judgement Rubric 662
PUPD but performing blood test
Calm, confident manner
Is visibly stressed/anxious and lacks confidence. Relies on client to make decisions and direct consultation.
e.g. Long silences and obvious uncertainty when deciding on treatment plan
Is tentative in the leader role; redirects some responsibility for decision making to the client. Moments of self-doubt, not 100% sure of treatment plan.
e.g. Offers treatment options but does not direct client/make
recommendation – client decides how to proceed
Is calm and confident in MOST situations. Directs the consultation but occasionally is unsure.
e.g. Changes mind about recommendations mid-consultation but otherwise confident and assumes responsibility for decision making
Assumes responsibility; is confident with
diagnosis/treatment plan.
e.g. Decides a treatment plan and relays this confidently to client
Clear Explanation
Explanations are
confusing and directions are unclear or
contradictory. Owners are confused.
e.g. Owner cannot make sense of instructions given
Explanations are mostly clear, though one element may cause confusion for the owner and need to be clarified.
E.g. Does not explain opinions clearly, owner has to ask questions to clarify
Explains carefully to clients and gives clear directions. The pace/tone may be inappropriate or may not check for owner understanding.
e.g. Explains plan well but speaks too quickly
Communicates at good pace; explains
interventions clearly; checks for understanding.
e.g. Explains plan at appropriate speed, clearly and checks for owner comprehension